Management of critically-ill patients (e.g., trauma and stroke) requires timely and responsive communication and care coordination between distributed care providers, i.e., emergency medical services (EMS) teams in the field and emergency care providers in the hospital. The fast-paced nature of emergency care, however, poses challenges to communication and care coordination during prehospital encounters, and may lead to confusion and miscommunication, and consequently, adversely affecting patient outcomes. The long-term goal of this project is to build an integrated computerized system that can support shared decision-making and care coordination between prehospital and hospital teams to improve patient outcomes.

Papers: iConference 2013, AMIA 2014, iConference 2015, CSCW 2017, CSCW 2018, JMS 2020, AMIA 2021, CSCW 2021, iConference 2022, CSCW 2022, JMIR Human Factors 2022, JAMIA Open 2022, JMIR Medical Informatics 2023, AMIA 2023, IJHCI2024.

Grants: NSF CRII (2020-2022, PI, $175,000), AHRQ R21 (2021-2023, PI, $300,000), NSF CAREER Award (2023-2028, PI, $500,000), NIH R15 (2024-2027, PI, $419,294).

With the advances in artifical intelligence (AI) technology in recent years, the AI technology has been increasingly used in a variety of healthcare domains, such as decision support, health education, and self-diagnoses. Despite its high potential, there are still many issues in the use and adoption of AI systems by clinicians and patients. For example, a key usability challenge of these systems is the lack of explanations of their reasoning to the user, making it difficult for users to understand systems’ hidden intelligence and determine when it is appropriate to trust the system-generated recommendation (e.g., medical advice). This challenge is exacerbated due to the scale and complexity of today’s machine learning and AI technologies. More transparent interface has been proven as a promising means of increasing user’s awareness of system actions and logics. In this research, the focus is to understand how to design trust-inspiring, transparent intelligent systems to support effective human-AI interaction and collaboration in healthcare settings.

Papers: ASIS&T 2020, JMIR 2020, CHI 2021, CSCW 2021, HIJ 2021, JMOS 2021, Frontiers in Computer Science 2023

Almost half of adults in the United States have limited health literacy, or limited capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions. Compared with those with adequate health literacy, patients with limited health literacy experience disparities in health care access and use of health information technology. Prior research has found that current technologies for managing and reporting personal health data, such as patient portals, do not provide enough, or the right, support to aid patient understanding their personal healthcare data (e.g., lab test results, clinical notes). It is unknown, however, how can we better support lay individuals with limited health literacy to read and comprehend their personal health data. To bridge this knowledge gap, this research will involve patients with limited health literacy and most at risk for poor health outcomes to develop mechanisms and novel technologies to promote the universal use of health information technology, and to aid them in managing, engaging with, and acting on personal health data.

Papers: MedInfo 2019, AMIA 2020, JMIR 2020, HIJ 2021, JMIR Human Factors 2021, Journal of Informatics for Health and Social Care 2021, AMIA 2021, JAMIA Open 2023, AMIA 2023, CHI2024.

Grants: AHRQ R21/R33 (2024-2029, co-I, $1,014,336)

The goal of this project is to design and develop an integrated information capture and display system to support trauma team awareness and decision making.

We conducted comprehensive user studies (e.g., observations, focus groups and interviews) to identify problems of current work practice and understand trauma teams' information needs. Findings from those studies specify the rules for display design. Display is designed and developed using an iterative and user-centered design approach that combines participatory design workshops, rapid prototyping, and heuristic evaluation during simulated events.

Collaborators: Diana Kusunoki, Nadir Weibel, Ivan Marsic, Genevieve Tuveson, Randall Burd

Papers: Group 2012, CSCW 2013, CHI 2014, JCSCW, ECSCW 2015

Paper checklists have become increasingly common in healthcare and are now used to support a wide range of complex medical activities. This project aims to create dynamic digital checklist to support emergency medical work.

We analyzed a set of trauma resuscitation videos to evaluate and quantify effects of paper-based checklist on team communication and interaction behaviors. We also reviewed and analyzed hundreds of paper checklists collected from real resuscitations to identify use patterns of paper-based checklist. The research findings were translated into design implications for digital checklist.

Papers: Group 2014, AMIA 2016